18 



MyxopKyceae 



heterocysts arc thicker than those of the ordinary vegetative cells and they 

 are composed of cellulose. 



Heterocysts are formed from young vegetative cells by the disintegration 

 of the incipient nucleus and the gradual assumption of a homogeneous 

 character by the whole of the cell-contents. The thickening of the wall 

 commences at the poles and gradually extends equatorially until the wall is 

 of equal thickness. There is a pore at one or both poles of the heterocyst, 

 depending upon its terminal or intercalary position, and surrounding the 

 pore is a slight thickening, which in older heterocysts may become 

 a minute plate and so close the pore. Within the heterocyst and im- 

 mediately adjacent to its pore or pores is a prominent granule, which 

 presents a very bright and refractive appearance, and consists, according 

 to Borzi, Hegler, and many recent investigators, of cyanophycin. Macallum 

 states that these granules give a reaction for masked iron and therefore 

 cannot consist of cyanophycin. Phillips also states that the heterocyst 

 is gradually filled with some substance passed into it from the other cells 

 through the pores by means of the protoplasmic threads which connect it 



Fig. 12. Successive stages in the formation of a heterocyst of Nnstoc sp. Greatly magnified 

 (after Phillips). Phillips states that some substance which stains very deeply with iron- 

 ammonia-alum hfematoxylin gradually fills up the cell ; this is represented black in the 

 figures. 



with adjacent cells of the filament. This substance stains deeply with iron- 

 ammonia-alum hsematoxylin and he suggests that it may possibly be related 

 to chromatin (see fig. 12). 



Heterocysts, except for a few instances, are normally solitary, but in 

 cultures and under unfavourable conditions they may become seriate. Brand 

 ('03), and subsequently Fritsch ('04) and others, have described the develop- 

 ment of an intercellular substance excreted during the formation of the 

 heterocysts. This substance, however, is not necessarily excreted by cells 

 which are being transformed into heterocysts, as it can be observed remote 

 from the heterocysts in Scytonema, and it is also excreted by the cells of 

 certain species of the OscillatoriaceaB, a family in which heterocysts do not 

 exist. 



The exact nature of heterocysts has long been a puzzle. They were 

 thought by Borzi (78) and Hansgirg ('87) to serve as limitations to the 

 length of the filaments, and they are sometimes, more especially in young 

 filaments and in cultures, concerned with the breaking of the filaments. 

 In this connection, it should be remembered, however, that in normal plants 



